"Even though we redesigned the website, I feel like our appearance in search results looks plainer than before." "Competitors have rich breadcrumbs and star ratings displayed, but we only have blue links and text." "Even if we search by company name, corporate information doesn't appear consolidated in the knowledge panel."—When taking over sites in client web development, we often receive consultations like these. The visual design is updated, yet appearance in search results—the "other front entrance"—has been left behind. In many cases, the cause is that structured data (JSON-LD) is missing, or even if included, its contents diverge from the actual content displayed on the page.
Structured data is a mechanism that communicates to search engines in a machine-readable format whether page content is an "article," "corporate information," or "breadcrumbs." When implemented correctly, rich results such as publication dates, authors, breadcrumbs, and company logos are more likely to appear in search results, increasing click-through rates. In this article, from our perspective handling client web development, we organize the types Google still supports today, proper syntax, and common implementation pitfalls.
Common Scenarios Where Structured Data Disappears During a Redesign
When structured data is missing from a site taken over, the cause is usually one of the following.
One is the case where what the old site's CMS output automatically disappeared during the migration. While a WordPress theme or plugin behind the scenes used to output Article schema for articles or BreadcrumbList schema for breadcrumbs, that output is simply dropped when moving to another CMS or static site. It is a state where only visual design was migrated, leaving hidden output behind.
The second is the case where it was never implemented in the first place. Although it looks polished visually, from a search engine's perspective it is "just HTML," lacking clues to determine whether it is an article or a product.
The third and most troublesome is the case where it is included, but the contents are incorrect. Author names remain empty from an old template, unrendered information is marked up, or broken JSON causes search engines to ignore it completely. Because people rest easy thinking "we already implemented it," this technical debt is hard to detect.
Appearance in search results is the entrance for organic traffic itself. The fundamentals of why structured data is effective for search and AI search in the first place are covered in our article on structured data and AI search.
JSON-LD Is the Only Format You Need
There are three formats for writing structured data—JSON-LD, Microdata, and RDFa—but when implementing afresh, choosing JSON-LD is definitely the right move. Google also recommends JSON-LD. The reason is clear: because JSON-LD is written together in a <script type="application/ld+json"> block, it can be completely decoupled from HTML visual markup.
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Organization",
"name": "グリームハブ株式会社",
"url": "https://gleamhub.net/",
"logo": "https://gleamhub.net/logo.png"
}
</script>Unlike Microdata, which embeds properties directly within HTML tags, JSON-LD allows you to manage structured data as an independent block. In client web development, you can add it later without breaking designs written by others, and updates only require touching a single place. The ease of maintenance is on another level. Note that our corporate site is built with Astro, and for structured data output we use astro-seo-schema to assemble JSON-LD by page type. Outputting it centrally at the layout level prevents omissions that occur when writing it manually for each article.
The Four Main Types You Should Include Today
While schema.org defines an enormous number of types, there are actually not many that are "worth implementing correctly first" for corporate sites and media handled in client web development. Narrowing down to the primary types Google treats as rich results makes decision-making much easier.
| Pattern | Target Pages | What Happens in Search Results |
|---|---|---|
| Organization | Entire site (corporate information) | Company name, logo, and social links are recognized more reliably by Google |
| Article | Articles, blogs, announcements | Publication date, author, and headline are conveyed structurally |
| BreadcrumbList | Hierarchical pages in general | Breadcrumb hierarchy is displayed in search results |
| FAQPage | Pages with frequently asked questions | (Display depends on environment; see cautions below) |
By correctly implementing just these three—Organization, BreadcrumbList, and Article on article pages—most of the symptoms mentioned at the beginning, such as "corporate info doesn't appear rich" or "breadcrumbs don't show up," will move toward resolution. If you are unsure about choosing types or setting priorities, thinking in this order will keep you on track. If you want to grasp the overall picture of SEO itself, our SEO beginner's guide also serves as a helpful foundation.
Do Not Overestimate FAQ or HowTo
Types like FAQPage and HowTo require caution. In recent years, Google has been streamlining rich results, significantly reducing rich presentations for FAQ and HowTo. Multiple reports indicate that following changes from 2025 into 2026, displays of FAQ rich results dropped drastically, and HowTo essentially ceased to be displayed. While the markup itself is not invalid, allocating person-hours under the assumption that "adding FAQ markup makes search results flashy" will not yield a matching return. In client web development, starting solidly with types that reliably work is the soundest approach.
The Biggest Mistake in Client Web Development Is Discrepancy With Rendered Content
More critical in client web development than technical syntax itself is that the marked-up content diverges from the content displayed on the page. Google's structured data guidelines explicitly prohibit marking up information not visible to users. The author name, date, and description written in JSON-LD must be visibly identical in the page body.
Violating this may lead, in the worst case, to manual actions that disqualify the site from rich results. While it does not necessarily lower search rankings directly, you lose eligibility for rich displays. Structured data added with great effort ends up having the opposite effect.
On one corporate site whose revamp we took over (company name withheld), legacy JSON-LD from an old template remained on the top page, still marking up the name of a former representative who had already resigned and services that were no longer offered. The displayed content conflicted with the JSON-LD contents. First, we reconciled the corporate info and services actually shown on the page against the JSON-LD, stripped everything not shown, and accurately rewrote only what was visible. What we did was not flashy work; we simply matched what is visible with what is conveyed to machines. That alone made the correct corporate information appear in the knowledge panel.
Beyond discrepancies with rendered content, common implementation mistakes follow several recurring patterns.
| Mistake | Symptom | How to Fix |
|---|---|---|
Forgetting to write @context | The entire block is ignored | Place "@context": "https://schema.org" at the beginning of each block |
| JSON syntax errors (trailing commas, etc.) | Structured data is not read at all | Remove trailing commas and unify quotes to double quotes |
| Missing required properties | Disqualified from rich results | Fill in required fields for each type (e.g., headline for Article) |
Because these are difficult to catch by visual inspection alone, relying on validation tools is essential.
Always Run Through Validation Tools Before Launch
Structured data is not finished once written; it is only complete once the process includes verifying whether machines can read it correctly before publishing. Verification relies primarily on two tools.
- Rich Results Test — Determines whether Google can recognize it as rich results and which types are valid. It distinguishes between errors (rich results will not appear) and warnings (missing optional fields, but it functions).
- Schema Markup Validator (validator.schema.org) — Validates syntactic correctness against schema.org vocabulary. You can confirm whether the structure is valid even for types not covered by Google.
In client web development, running both tools to confirm "zero errors" prior to delivery is bundled into the scope. While warnings may sometimes be acceptable, remaining errors prevent your markup from being reflected in search results. Even after launch, monitoring the error count of detected structured data continuously in Google Search Console's enhancement reports helps you catch issues quickly when CMS updates break markup.
Alongside appearance in search results, page experience factors like loading speed impact clicks and evaluations. When optimizing together, reviewing perspectives from our Core Web Vitals article simultaneously avoids wasted effort.
Where to begin
If you want to recover search result visibility for a site you took over, start by passing current pages through validation tools one by one to determine whether structured data is present and, if present, whether it matches displayed content. If nothing is included, properly add Organization, BreadcrumbList, and Article on article pages. If outdated, erroneous markup remains, adjust it to match displayed content, or remove it entirely before re-adding. It is not flashy, but it reliably brings the appearance of your search result entrance on par with competitors.
As a first move, we recommend achieving zero errors in the Rich Results Test for Organization on the top page and Article and BreadcrumbList on article listings. Tidying up just this part changes the display of company info and breadcrumbs, making the benefits easy to experience.
If you are facing challenges such as search result appearance falling behind competitors after a redesign, not knowing whether structured data exists on a site you took over, or rich results failing to appear despite adding markup, please contact GleamHub. We take stock of structured data on existing sites using validation tools, isolate discrepancies with rendered content, and correctly reimplement starting with the types that deliver real impact.
Sources
- Intro to How Structured Data Markup Works － Google Search Central
- General Structured Data Guidelines － Google Search Central
- Learn About Article Schema Markup － Google Search Central
- JSON-LD Tutorial: Syntax, @id, @graph, Validation (2026)
- Schema Markup After March 2026: Structured Data Update － DigitalApplied
- A beginners guide to JSON-LD Schema for SEOs － SALT.agency
